Nephrectomy remains the standard of care for patients

Transcription

1 ORIGINAL ARTICLE Non-neoplastic Renal Diseases are Often Unrecognized in Adult Tumor Nephrectomy Specimens A Review of 246 Cases Kammi J. Henriksen, MD, Shane M. Meehan, MB, BCh, and Anthony Chang, MD Abstract: The pathologic evaluation of tumor nephrectomy specimens focuses on the diagnosis, grading, and staging of the neoplasm. The presence of coincidental non-neoplastic diseases in these specimens may have significant implications for patient outcomes. The purpose of this study is to determine the spectrum of non-neoplastic disease processes that may be overlooked in tumor nephrectomies, and to ascertain the extent to which surgical pathologists are trained to recognize these lesions. We reviewed the hematoxylin and eosin-stained slides of 246 adult tumor nephrectomy specimens with an emphasis on the non-neoplastic renal parenchyma. Further analysis of cases with pathologic alterations included special stains and direct immunofluorescence microscopy. The surgical pathology reports were reviewed to determine whether the non-neoplastic lesions were originally identified. We also surveyed United States pathology residency programs to determine how many require training in medical renal pathology. Forty-one cases (16.7%) had alterations, such as diffuse and/or nodular mesangial sclerosis, mesangial hypercellularity, or glomerular basement membrane thickening that warranted further study. After further work-up and clinical correlation, the pathologic changes in 24 cases were categorized as follows: diabetic nephropathy (19 cases) of which one demonstrated atheroembolic disease, thrombotic microangiopathy (3 cases), sickle cell nephropathy (1 case), and focal segmental glomerulosclerosis (1 case). Twenty-one (88%) of these diagnoses were not identified at initial pathologic evaluation. Only 35 of 98 pathology residency programs that responded to our survey require any formal training in medical renal pathology. Although accurate pathologic evaluation of renal neoplasms remains essential, surgical pathologists should be aware that coincidental nonneoplastic renal diseases are common, often not recognized, and may have important implications for patient care. Further consideration should be given to the training requirements of pathology residents and the guidelines for evaluation of nephrectomy specimens to avoid missing non-neoplastic renal lesions. From the Department of Pathology, The University of Chicago, Chicago, IL. Reprints: Anthony Chang, MD, Department of Pathology, The University of Chicago, 5841 S. Maryland Avenue, Room S-628 (MC6101), Chicago, IL ( anthony.chang@uchospitals.edu). Copyright r 2007 by Lippincott Williams & Wilkins Key Words: renal neoplasm, nephrectomy, medical renal pathology, diabetic nephropathy (Am J Surg Pathol 2007;31: ) Nephrectomy remains the standard of care for patients with a suspected renal mass despite studies that have established nephrectomy as an independent risk factor for developing chronic renal insufficiency. 11,13 It has been reported that 22.4% and 11.6% of total and partial nephrectomy patients, respectively, demonstrate significant elevation of their serum creatinine 10 years after surgery due to the significant loss of nephrons. 12 Previous studies have established that glomerular disease 3 or arteriolonephrosclerosis 6 frequently coexist with renal cell carcinoma. Bijol et al 4 recently found that over 60% of tumor nephrectomy specimens have coincidental medical renal diseases, of which the presence of diabetic nodular glomerulosclerosis or >20% global glomerulosclerosis was predictive of a significant decline in renal function within 6 months of surgery. Thus, the accurate pathologic evaluation of the non-neoplastic renal parenchyma in nephrectomy specimens is important for subsequent patient management. Considering that the evaluation of nephrectomy specimens primarily focuses on the diagnosis, grading, staging, and margin status of the tumor in question, we hypothesized that nonneoplastic lesions might easily be overlooked during the routine pathologic evaluation of these specimens. The purpose of this study is to determine the spectrum of diagnostic discrepancies in the non-neoplastic renal parenchyma that may be observed and to determine whether surgical pathologists receive sufficient training to recognize these lesions. MATERIALS AND METHODS We reviewed the files at the University of Chicago department of pathology and identified 246 adult partial or total nephrectomy specimens between January 1997 and January 2003, which were performed with a clinical suspicion of a mass lesion. Exclusion criteria included renal allografts, specimens with findings consistent with end-stage renal disease and partial nephrectomies with <0.5 cm of surrounding non-neoplastic parenchyma for evaluation. The hematoxylin and eosin (H&E)-stained Am J Surg Pathol Volume 31, Number 11, November

2 Henriksen et al Am J Surg Pathol Volume 31, Number 11, November 2007 slides were reviewed by 2 observers (K.J.H. and A.C.) without prior knowledge of clinical information and with an emphasis on the non-neoplastic renal parenchyma. The histologic identification of glomerular alterations, such as diffuse and/or nodular mesangial sclerosis, mesangial hypercellularity (defined as more than 3 cell nuclei per mesangial area on a 4-mm thick tissue section, away from the vascular pole), 2 or glomerular basement membrane thickening, triggered additional evaluation of these sections by periodic acid-schiff (PAS) stains. After the identification of abnormal glomerular alterations in the PAS-stained tissue sections, the computerized patient records of these cases were searched for pertinent clinical information. Immunofluorescence (IF) microscopy of IgG, IgA, and k and l light chains on paraffin tissue sections with appropriate controls was performed on a subset of cases with significant glomerular alterations using the following protocol. Briefly, 4-mm sections were cut from the original paraffin blocks and deparaffinized with 0.25% trypsin-ethylene diaminetetra acetic acid for 1.5 hours at 371C. After washing, they were placed in phosphate-buffered saline and stained with fluorescein isothiocyanate-conjugated antibodies to IgG, IgA, or k or l light chains (DAKO, Carpinteria, CA) for 1 to 3 hours. Congo red stains were performed on 8-mm sections from the original paraffin blocks for a subset of cases with nodular mesangial sclerosis. Standard immunohistochemistry techniques were used for CD61 (DAKO) staining of one specimen. Severe diabetic nephropathy was defined as nodular glomerulosclerosis involving >50% of the glomeruli, and the remaining cases were classified as mild to moderate. Varying degrees of global glomerulosclerosis, interstitial fibrosis, tubular atrophy, and arteriosclerosis were considered nonspecific findings in this study. We reviewed the original surgical pathology reports to determine whether the non-neoplastic renal lesions were initially identified. We also conducted a survey of all pathology residency program directors by . When a response was not obtained from the program director, an attempt to contact the renal pathologist and/or chief resident at the institution was made. We specifically asked whether each residency program offered a required or elective rotation in medical renal pathology and the duration of this rotation. RESULTS Of the 246 patients (151 males and 95 females) from 175 total and 71 partial nephrectomies, the average age was 60 years (range: 28 to 87 y). The pathologic diagnoses are summarized in Table 1. After review of the H&E slides, we identified 41 (16.7%) of 246 cases with significant glomerular alterations. After additional review of the PAS stains, 17 cases failed to show a definitive pathologic disease beyond mild mesangial hypercellularity, matrix expansion, or mild arterionephrosclerosis. Of note, 2 patients had hepatitis C infection, 13 had hypertension, and 1 was diabetic. TABLE 1. Summary of Pathologic Diagnoses Diagnosis No. Cases (%) No. Cases With Non-neoplastic Renal Disease Renal cell 123 (50) clear cell Renal cell 30 (12) 7 23 papillary Renal cell 26 (11) 2 24 unspecified Renal cell 5 (2.0) 0 5 sarcomatoid Renal cell 4 (1.6) 0 4 chromophobe Oncocytoma 21 (8.5) 2 19 Angiomyolipoma 11 (4.5) 0 11 Transitional cell 10 (4.1) 1 9 carcinoma 9 (3.7) 0 9 Simple and/or multilocular cysts Metastatic ovarian serous adenocarcinoma Other disease process* No. Cases With Nondiagnostic Parenchymal Alterations 1 (0.4) (2.4) 2 4 *Includes: xanthogranulomatous pyelonephritis, chronic pyelonephritis, nephrocalcinosis, hematoma, and pyohydronephrosis. Although the constellation of findings in some of these cases might be consistent with mild arterionephrosclerosis, this entity was not considered a specific diagnostic category for this study, because it was difficult to determine whether there was truly a diagnostic discrepancy when one or more of the histologic findings was mentioned descriptively in the original surgical pathology reports. The remaining 24 cases showed pathologic renal disease classifiable as follows: diabetic nephropathy, thrombotic microangiopathy (TMA), sickle cell nephropathy, and focal segmental glomerulosclerosis. A review of the original surgical pathology reports of these 24 cases found that only 3 of the severe diabetic nephropathy cases were originally identified. Non-neoplastic renal diseases in the remaining 21 cases (88%) demonstrating definite pathologic changes were not identified during the initial evaluation. Nineteen cases had diffuse and/or nodular mesangial sclerosis with varying degrees of basement membrane thickening, hyalinosis lesions (capsular drops or fibrin caps), mesangiolysis, and arteriolar hyalinosis (Figs. 1A, B). Twelve of these cases were classified as severe diabetic nephropathy and the remaining 7 as mild to moderate. An organized atheroembolus with a cholesterol cleft was noted in a small artery of one mild to moderate diabetic nephropathy case. Given that the differential diagnosis of nodular mesangial sclerosis includes amyloidosis, 1704 r 2007 Lippincott Williams & Wilkins

3 Am J Surg Pathol Volume 31, Number 11, November 2007 Non-neoplastic Renal Diseases in Tumor Nephrectomies FIGURE 1. A, This glomerulus demonstrates mild, diffuse mesangial sclerosis in a case of mild to moderate diabetic nephropathy (PAS, 400 ). B, Nodular mesangial sclerosis characterized the glomeruli in cases of severe diabetic nephropathy (PAS, 400 ). monoclonal immunoglobulin deposition disease (light and/or heavy chain deposition disease), fibrillary glomerulonephritis, and immunotactoid glomerulopathy, we attempted to exclude these other diagnostic possibilities. None of the 19 cases showed positive Congo red staining, which excludes the diagnosis of amyloidosis. IF microscopy for IgG, IgA, and k and l light chains showed no significant glomerular, tubular basement membrane, or interstitial staining, which argues against an immune complex-mediated disease process or monoclonal immunoglobulin deposition disease. Eighteen of 19 patients had a history of either type 1 or 2 diabetes mellitus. We included 1 case without the clinical history of diabetes in this diagnostic category, because the presence of diffuse and focally nodular mesangial sclerosis with fibrin caps and arteriolar hyalinosis was most suggestive of diabetic nephropathy. Three cases demonstrated findings of TMA. One case showed several glomerular capillaries distended by thrombi in a small cortical region (Fig. 2A). CD61 immunohistochemical staining highlighted these capillary thrombi. The other 2 TMA cases revealed double contours and duplicated glomerular basement membranes (Fig. 2B) involving several glomerular capillaries FIGURE 2. A, A large thrombus (arrows) is present in the lumen of the hilar arteriole in this patient with acute TMA (H&E, 400 ). B, This glomerulus with chronic TMA showed duplication of the glomerular basement membranes (arrow) (PAS, 400 ). r 2007 Lippincott Williams & Wilkins 1705

4 Henriksen et al Am J Surg Pathol Volume 31, Number 11, November 2007 FIGURE 3. Focal segmental glomerular scarring and glomerular basement membrane duplication (arrows) was present in this case of sickle cell nephropathy (PAS, 400 ). on PAS stain, which were findings consistent with a chronic thrombotic microangiopathic injury. One case demonstrated focally marked mesangiolysis. All 3 cases showed no significant IF staining for IgG, IgA, and k and l light chains. Other common causes of TMA, including thrombotic thrombocytopenic purpura, hemolytic uremic syndrome, antiphospholipid antibody syndrome, scleroderma, or radiation therapy, were not necessarily excluded due to the limited clinical information that was available. Only a vague history of vasculitis and hypertension was given in the patient with acute TMA. Of the 2 chronic TMA cases, one had hepatitis C infection, and the other had a history of recurrent episodes of malignant hypertension and chronic renal insufficiency not otherwise specified. A case of sickle cell nephropathy in a 47-year-old man was confirmed after further investigation of the patient s clinical history. Of 50 glomeruli available for examination, 35% were globally sclerotic and 15% showed segmental sclerosis with segmental accumulation of matrix, obliteration of capillary loops, and fibrous attachments to Bowman s capsules (Fig. 3). Occasional double contours of the glomerular basement membranes were noted in some glomerular capillaries. Prominent hemosiderosis was found in many tubular epithelial cells. There was no significant IF staining for IgG, IgA, and k and l light chains. The presence of rodlike inclusions in the red blood cells consistent with polymerized hemoglobin was observed by electron microscopy (EM). Focal segmental glomerulosclerosis was identified in a patient with a concomitant ovarian papillary serous cystadenocarcinoma at the time of nephrectomy and a clinical history of congenital ureteropelvic junction obstruction and recurrent pyelonephritis. We were unable to determine whether the patient had significant proteinuria. There was prominent dilatation of the renal pelvis consistent with hydronephrosis. A portion of renal cortex immediately adjacent to this region showed prominent interstitial fibrosis and tubular atrophy with very few residual globally sclerotic glomeruli. The renal cortex removed from the dilated calyx showed no significant interstitial fibrosis or tubular atrophy. Approximately, 10% of the glomeruli were globally sclerotic. Two percent of the glomeruli showed segmental sclerosis characterized by the presence of segmental accumulation of matrix, obliteration of a few capillaries by hyaline material and foam cells and fibrous attachments to Bowman s capsules. IF microscopy for IgG, IgA, and k and l light chains was negative. The arteries revealed mild intimal fibrosis. The arterioles had diffuse and focally prominent subendothelial hyalinosis. We conducted a survey of all pathology residency training programs in the United States to determine the percentage of programs that offer formal training in renal pathology. Of 147 US pathology residency programs, 98 responses (67%) were received. Thirty-four programs (36%) require residents to complete a rotation in renal pathology, which generally lasted 1 month, but ranged from 1 week to 3 months. Seven of these 35 programs combined renal with other specialties (neuropathology, dermatopathology, hepatic, pediatric, and/or ophthalmic pathology). Of the 63 programs that do not require a rotation in renal pathology, 35 reported that they do offer an elective rotation and 2 of the programs are in the process of initiating a mandatory rotation. Thus, 28 (29%) pathology residency programs do not currently offer any formal training in medical renal pathology. DISCUSSION This is the first study to focus on diagnostic errors of non-neoplastic renal diseases in tumor nephrectomy specimens. Although this study was conducted at a single institution, our findings highlight an important diagnostic issue that is pertinent to all practicing surgical pathologists. Diagnostic errors in surgical pathology can be classified into the following categories: preanalytic, analytic, and postanalytic. 16 Preanalytic errors are primarily made by submitting physicians, such as obtaining tissue from the wrong patient or providing incorrect clinical information. Analytic errors include specimen mix-ups, slide label errors, specimen contamination, diagnostic errors, etc. Postanalytic errors include miscommunication of the diagnosis, untimely delays, or absence of reports for clinicians. Major errors would have significant therapeutic and/or prognostic implications, whereas minor errors would have no significant effect on the patient. Without the availability of significant clinical follow-up in our patients, we could not categorize the diagnostic errors in our study into a major or minor category with any degree of certainty r 2007 Lippincott Williams & Wilkins

5 Am J Surg Pathol Volume 31, Number 11, November 2007 Non-neoplastic Renal Diseases in Tumor Nephrectomies It is important to understand how a surgical pathologist might commit a diagnostic error and overlook the presence of a non-neoplastic renal lesion in a nephrectomy specimen. With the primary focus on the neoplasm in question, the pathologist might neglect to evaluate the non-neoplastic renal parenchyma altogether or fail to identify a lesion due to a cursory evaluation or fail to recognize a lesion despite careful evaluation. Although we did not analyze this specific issue, all of these possibilities likely contributed to the diagnostic discrepancies that were found in our study. The College of American Pathologists (CAP) and the Association of Directors of Anatomic and Surgical Pathology (ADASP) have recently established cancer protocols and checklists to standardize the pathologic evaluation of various tumor specimens and minimize the possibility of omitting important pathologic features. 1,7 Although these measures may improve and ensure systematic evaluation of the neoplasm, the section regarding evaluation of the non-neoplastic parenchyma is currently optional for both checklists. Additionally, although the CAP protocol specifically suggests scanning for glomerular disease, the ADASP protocol fails to mention that compartment. Thus, the specific renal diseases that we encountered in our study would likely remain undiagnosed in this new era of synoptic reporting. It is important to systematically evaluate the anatomic compartments of the non-neoplastic renal parenchyma: glomeruli, tubules, interstitium, and vessels. The focality of some lesions cannot be overstated. For example, a few of the mild-moderate diabetic nephropathy cases showed only an occasional Kimmelstiel- Wilson type nodule after inspection of over 30 glomeruli. The diabetic patients in our study had not been previously biopsied, so the surgical pathologist would have been the first person to identify the presence of diabetic nephropathy. The diagnosis of mild diabetic nephropathy is significant because this presents an early opportunity to control the diabetic injury process. The presence of mild diabetic nephropathy would also have a greater impact on a patient with only 1 remaining kidney versus someone with 2 kidneys. In type 1 diabetic patients, it has been shown that achievement of normoglycemia after pancreas islet cell transplantation can lead to reduction of nodular mesangial sclerosis that is characteristic of diabetic nephropathy. 5,10 It has also been demonstrated that aggressive systolic blood pressure control and blockade of the renin-angiotensin system is renoprotective in type 2 diabetics. 15 Thus, similar to the treatment of cancer, early detection of renal diseases is important for proper management of such diseases. The surgical pathologist may know that the average 5-year survival rate for renal cell carcinoma patients is >90%, 75% to 90%, and 59% to 70% for those with stages 1, 2, and 3 disease, respectively. 17 In comparison, the majority of elderly patients with end-stage renal disease being treated with dialysis will not live longer than 5 years. The 2004 US Renal Data System shows that the average 60, 70, and 80-year old end-stage renal disease patient on dialysis has a life expectancy of 4.3, 3.1, and 2.2 years, respectively. 9 On the basis of these statistics, a patient with both stage 3 renal cell carcinoma and endstage renal disease will be more likely to succumb to their end-stage renal disease than their renal cell carcinoma. Thus, especially if one is dealing with a benign neoplasm, such as an angiomyolipoma or oncocytoma, the status of the non-neoplastic renal parenchyma is the most important determinant for the patient s outcome. Only 2 studies in the medical literature have systematically evaluated the non-neoplastic renal parenchyma in tumor nephrectomy specimens. Beaufils et al 3 reviewed 40 cases of renal cell carcinoma and found significant mesangial sclerosis with or without hypercellularity in 7 cases of which 2 had mild proteinuria and notable IF staining for C3 and/or IgM in 13 patients and IgA and C3 in 1 patient. Bijol et al 4 evaluated the nonneoplastic parenchyma in 110 tumor nephrectomy specimens. Sixty of these specimens were carefully evaluated prospectively by light, IF, and EM. This study establishes a baseline of non-neoplastic renal diseases that the surgical pathologist can expect to encounter in such specimens. They reported that >60% of 110 cases demonstrated apparent pathologic abnormalities. Roughly one-half of these cases demonstrated diabetesrelated changes, whereas the other half exhibited marked vascular changes and parenchymal scarring. Correlation of these findings with subsequent clinical data revealed that patients with severe histologic findings demonstrated a significant rise in serum creatinine by 6 months postnephrectomy as compared to those patients with unremarkable renal parenchyma, indicative of an accelerated decline of renal function. In comparison to the study by Bijol et al, our study identified significant morphologic abnormalities using primarily H&E-stained tissue sections with a focus on diagnostic discrepancies of the non-neoplastic parenchyma. Although 50 specimens from their study were performed retrospectively, the presence of diagnostic discrepancies was not mentioned. Using our approach, we may have overlooked some lesions, such as early membranous or IgA nephropathy, which may show minimal glomerular alterations that would have not been identified by H&E or PAS staining alone. Some of these lesions would perhaps only be detectable with the use of IF and/or EM. Of note, we were limited to staining paraffin-embedded tissue sections, which has a lower sensitivity compared with standard IF microscopy performed on frozen tissue sections. 14 Additionally, some of our diagnostic criteria differed from Bijol et al, so that we could establish whether a clear diagnostic discrepancy was present. For this reason, we did not consider cases with findings of severe parenchymal scarring or end-stage renal disease as diagnostic entities, because it was difficult to determine if a diagnostic discrepancy existed if some histologic features were mentioned descriptively in the original surgical pathology report. We also chose not to include glomerular hypertrophy as an early lesion of diabetic nephropathy. r 2007 Lippincott Williams & Wilkins 1707

6 Henriksen et al Am J Surg Pathol Volume 31, Number 11, November 2007 One might argue that the renal parenchyma of all tumor nephrectomy specimens should undergo the same evaluation that is given to medical renal biopsies. However, the costs and benefits of such a course of action would have to be considered, which is beyond the scope of our current study. As previously mentioned, no standard guidelines have been established for the proper evaluation of the non-neoplastic renal parenchyma of tumor nephrectomy specimens. To reduce the potential for diagnostic errors, we recommend that a PAS and/or Jones methenamine silver stain should be ordered on a representative section of the non-neoplastic renal parenchyma for all tumor nephrectomy specimens. The special stains should be available for the initial evaluation when the H&E slides are reviewed. This is a simple solution that would remind the pathologist to evaluate the renal parenchyma in a systematic fashion. In particular, the glomerular and vascular compartments should be carefully evaluated. The identification of abnormal histologic features, such as glomerular hypercellularity, mesangial sclerosis, or basement membrane thickening, should result in obtaining additional clinical information and IF and/or EM. Both IF and EM can be performed on formalin-fixed, paraffin-embedded tissue sections, albeit with some decreased sensitivity in the former and processing/preservation artifact for the latter. 8,14 Finally, we recommend that future revisions of the CAP and ADASP cancer protocols and checklists require the routine evaluation of the non-neoplastic renal parenchyma in tumor nephrectomy specimens. Our survey of pathology residency programs shows that only 36% have a mandatory requirement in medical renal pathology. Of interest, 29% of residency programs offer no formal training in medical renal pathology, and this percentage is likely higher due to a response bias to our survey, considering that most of the residency programs that did not respond our survey are smaller in size and may not be able to establish a required or elective rotation in medical renal pathology. Although there is no previous data with which to compare our findings, the formal training of pathology residents in medical renal pathology has probably not changed significantly over the years. Thus, a significant number of surgical pathologists may not be sufficiently trained to assess the nonneoplastic renal parenchyma of tumor nephrectomy specimens. The standard training requirements of the anatomic pathology resident with regard to medical renal pathology deserves further consideration. In summary, our study shows that nearly one of every 10 tumor nephrectomy specimens demonstrates a non-neoplastic renal disease. All surgical pathologists should be aware that non-neoplastic renal lesions are common and often overlooked in tumor nephrectomy specimens. A PAS and/or Jones methenamine silver stain should be routinely used in the evaluation of the nonneoplastic renal parenchyma of every tumor nephrectomy specimens. In some instances, the accurate evaluation of the non-neoplastic renal parenchyma can be as important as the classification of the renal neoplasm. REFERENCES 1. ADASP Checklists and guidelines for surgical pathology reports of malignant neoplasms [Association of Directors of Anatomic and Surgical Pathology Web site]. Updated November Available at: Accessed January 18, D Agati VD, Jennette JC, Silva FG. Non-neoplastic Kidney Diseases. Washington, DC: American Registry of Pathology; 2005: Beaufils H, Patte R, Aubert P, et al. Renal immunopathology in renal cell carcinoma. Virchows Arch Pathol Anat. 1984;404: Bijol V, Mendez GP, Hurwitz S, et al. Evaluation of the nonneoplastic pathology in tumor nephrectomy specimens: predicting the risk of progressive renal failure. Am J Surg Pathol. 2006; 30: Bilous RW, Mauer SM, Sutherland DE, et al. The effects of pancreas transplantation on the glomerular structure of renal allografts in patients with insulin-dependent diabetes. N Engl J Med. 1989;321: Budin RE, McDonnell PJ. Renal cell neoplasms: their relationship to arteriolonephrosclerosis. Arch Pathol Lab Med. 1984;108: Cancer protocols and checklist [College of American Pathologists Web site]. Updated November 16, Available at: Accessed January 18, Collar J, Cattell V. Paraffin-processed material is unsuitable for diagnosis of thin-membrane disease. Nephron. 1995;69: Collins AJ, Kasiske B, Herzog C, et al. Excerpts from the United States Renal Data System 2004 annual data report: atlas of endstage renal disease in the United States. Am J Kidney Dis. 2005;45: S Fioretto P, Bruseghin M, Berto I, et al. Renal protection in diabetes: role of glycemic control. J Am Soc Nephrol. 2006;17:S86 S Huang WC, Levey AS, Serio AM, et al. Chronic kidney disease after nephrectomy in patients with renal cortical tumors: a retrospective cohort study. Lancet Oncol. 2006;7: Lau WK, Blute ML, Weaver AL, et al. Matched comparison of radical nephrectomy vs nephron-sparing surgery in patients with unilateral renal cell carcinoma and a normal contralateral kidney. Mayo Clin Proc. 2000;75: Leibovich BC, Blute ML. Surgical management of renal cell carcinoma. Semin Oncol. 2006;33: Nasr SH, Galgano SJ, Markowitz GS, et al. Immunofluorescence on pronase-digested paraffin sections: a valuable salvage technique for renal biopsies. Kidney Int. 2006;70: Pohl MA, Blumenthal S, Cordonnier DJ, et al. Independent and additive impact of blood pressure control and angiotensin II receptor blockade on renal outcomes in the irbesartan diabetic nephropathy trial: clinical implications and limitations. J Am Soc Nephrol. 2005;16: Sirota RL. Defining error in anatomic pathology. Arch Pathol Lab Med. 2006;130: Tsui KH, Shvarts O, Smith RB, et al. Prognostic indicators for renal cell carcinoma: a multivariate analysis of 643 patients using the revised 1997 TNM staging criteria. J Urol. 2000;163: r 2007 Lippincott Williams & Wilkins